Massively parallel, targeted DNA resequencing and cardiovascular disease

Hypertrophic cardiomyopathy (HCM) and dilated cardiomyopathy (DCM) are important causes of sudden death and heart failure. These cardiomyopathies are often caused by single, rare DNA mutations. Identification of such mutations in subjects enables earlier diagnosis and therapeutic intervention. However, conventional sequencing does not find mutations in many cardiomyopathy subjects.;The development of massively parallel DNA sequencing (MPS) technologies has dramatically decreased the cost and increased the yield of DNA sequence analyses. This thesis presents a new approach to sequencing specific genomic regions, filter-based hybridization capture followed by MPS. This strategy was applied to the study of 115 kilobases of the human genome encompassing 47 genes. Sequencing of captured subgenomic libraries interrogated 99.8% of targeted bases ≥20 times (∼40,000-fold enrichment), enabling sensitive and specific detection of single-nucleotide and copy-number variation. This study identified an 11 kilobase tandem insertion of MYBPC3 in an HCM subject, suggesting that complex genomic rearrangements may cause HCM. Further application of this targeted resequencing methodology should help better define the genetic architectures of disease.;TTN, the gene encoding the sarcomere protein titin, has been strongly implicated in cardiomyopathies, but has been incompletely studied due to the monumental size of its coding sequence (∼100 kilobases). This thesis describes the sequencing of TTN in a large cohort of DCM, HCM, and control subjects using filter-based hybridization capture followed by MPS or traditional dideoxy sequencing. We identified 58 structural mutations (22 nonsense, 20 frameshift, 15 splicing, and 1 large tandem insertion) of full-length titin. These mutations were enriched in DCM subjects as compared to HCM subjects (P=2x10-11) or to control subjects (P=7x10 -4) and were co-inherited with DCM in families (LOD=6.5). The distribution of these mutations across titin was nonrandom (P=0.006), suggestive of a dominant-negative mechanism. We estimate that the prevalence of TTN structural mutation in idiopathic DCM is 10%--30%, and is therefore the most common known genetic cause of DCM. Incorporation of massively parallel sequence analyses of TTN into clinical genetic screens should increase the detection of pathogenic mutations by at least 50%, enabling earlier diagnosis and interventions to prevent disease progression. Follow-up mechanistic and clinical studies should improve our understanding of cardiomyopathy pathophysiology and may enable more specific clinical prognostication and management.